NA rURE 



[April 2S, 1887 



magnitude and direction of movement, and their relation to areas 

 of low pressure. The latter subject is illustrated with a plate 

 giving the isobars for December 15, 1882, in the northern hemi- 

 sphere, and it is shown generally that the movement of areas of 

 high pressure depends upon very different causes from that of areas 

 of low pressure, which seem to be endowed with a power of loco- 

 motion residing within themselves. Areas of high pressure 

 exhibit no such power, their movement seeming to depend 

 entirely on external forces. — The faults of South- West Virginia, 

 by John J. Stevenson. Here is given a summary of the infor- 

 mation communicated by the author in several memoirs read 

 before the American Philosophical Society (1880-84) on a 

 reconnaissance made by him of the faulted region in Virginia 

 from the Tennessee line to nearly twenty miles beyond New 

 River, a total distance of 150 miles. — On Taconic rocks and strati- 

 graphy, with a geological map of the Taconic regions, part 2, 

 by James D. Dana. The general facts are here detailed which 

 bear on the geographical distribution of the limestone and other 

 rocks. A description is given of Williamstown, regarded as the 

 bitthplace of the Taconic system, and it is shown generally that 

 the great quartzite formation, forming the foundation of the 

 Palreozoic of the region, derived its material from ArchEeair 

 formations of the vicinity and not from the fabled "Atlantis" as 

 some geologists have supposed. — Irish esker drift, by G. H. 

 Kinahan. It is pointed out that, in common with some other 

 observers. Prof. Carvill Len is confounds true esker drift and 

 ridges of esker-like drift. His statement in the December 

 number of the Amcricati Journal of Science that the Irish 

 eskers appear to be adjuncts of the melting of the ice-sheets is 

 shown to be impossible. — Physical characteristics of the northern 

 and north-wes!ern lakes, by L. Y. Schermerhorn. The results 

 are given of the lately completed surveys made by the United 

 States of all the great lakes draining through the St. Lawrence 

 to the Atlantic. The total water-area is 95,275, and the total 

 area of the lacustrine basin over 270,000, square miles. The 

 length of shore-line with connecting rivers is about 5400 miles, 

 the extreme depth of Superior looS feet, or over 406 below 

 sea-level, the mean annual rainfall of the whole basin 31 

 inches, the volume of water in the lakes about 6000 cubic miles, 

 and the discharge of Ontario at St. Lawrence River 300,000 

 cubic feet per second. — Mineralogical notes from the laboratory 

 of Hamilton College, by Albert H. Chester. Specimens are 

 described and analysed of fuchsite, pink celestite, zinkenite, 

 brochantite, pectolite, crystals of barite, scorodite, and bis- 

 muthite. — The topography and geology of the Cross Timbers and 

 surrounding regions in Northern Texas, by Robert T. Hill. This 

 wooded zone penetrating in two belts from Indian territory 

 through the surrounding prairie southwards to 32° N. lat. , is ex- 

 plained by the detritus of arenaceous strata which occupy well- 

 defined horizons in the geological series, and which have been 

 exposed by the denudation of the overlying strata. — American 

 Jurassic mammals, by Prof. O. C. Marsh. In tliis paper are 

 described the remains of several hundred individuals which have 

 come to light since the appearance of the author's previous 

 papers on the subject. Although fragmentary, the remains are 

 usually well preserved, comprising the lower jaws, teeth in situ, 

 various portions of the skull, vertebra;, and other parts of the 

 skeleton. Placental as well as marsupial mammals occur in the 

 oldest formations, whence the inference that the former do not 

 derive from the latter evolutionally, as is supposed, but that 

 both of these orders descend in independent lines from a common 

 ancestor. 



SOCIETIES AND ACADEMIES 



London 



Royal Society, March 24. — " On the Magnetisation of 

 Iron in Strong Fields." By Prof. J. A. Ewing, University 

 College, Dundee, and William Low. 



The behaviour of iron and steel when subjected to very strong 

 magnetising forces is a matter of considerable practical and very 

 great theoretical interest, especially from its bearing on the 

 molecular theory of magnetisation, which assigns an upper 

 limit to the intensity of magnetism that a piece of iron can 

 acquire, and even suggests that the metal may become diamag- 

 netic under the influence of a sufficiently great force. All ex- 

 periments hitherto made, by magnetising iron in the field of an 

 electric solenoid, have shown that the intensity of magnetism, jj, 

 as well as the induction, 13, is increasing with the highest values 



actually given to the magnetising force, 'lif). It is scarcely 

 practicable, however, to produce by the direct action of a 

 magnetising solenoid, a field'whose force exceeds a few hundreds 

 of C.G. S. units. 



In the space between the pole-pieces of a strong electro- 

 magnet we have a field of force of much greater intensity than 

 it is practicable to produce by the direct action of the electric 

 current. This field is not well adapted for experiments whose 

 object is to determine with precision Ihe relation of magnetisa- 

 tion to magnetising force, on account of the distortion which it 

 undergoes when the piece of iron to be magnetised is introduced 

 into it. It is, however, well suited for e.xperiments whose object 

 is to determine how much magnetism the metal can be forced to 

 take up. 



In the authors' experiments, bobbins of Lowmoorand Swedish 

 wrought-iron and cast-iron were magnetised by placing them 

 between the pole-pieces of a large electro-magnet. The bobbins 

 consisted of a short narrow central neck with conical ends. 

 The magnetic induction in the neck was measured ballistically 

 by means of an induction coil, consisting of a single layer of fine 

 wire, wound on the neck ; and the magnetic field in the air- 

 space immediately contiguous to the neck was also measured by 

 means of a second induction coil wound over the first, and of 

 slightly greater diameter. This enabled the non-ferrous space 

 under the inner induction coil to be corrected for, and also gave 

 an approximation to the v.alue of ftj, the magnetic force acting on 

 the metal. The magnetic force varied up to about 11,000 

 C.G. S. units, and the highest induction observed (in a sample 

 of Lowmoor wrought-iron) was 32,880 C.G.S. units. In some 

 instances the magnetic induction was observed by withdrawing 

 the bobbin ; in others the bobbin was turned round suddenly so 

 that its magnetism was reversed. The following results refer to 

 Lowmoor wrought-iron and to cast-iron. 



Lowmoor wrotight-iron 



per 



, induction in field magnets, 

 the metal amperes 



3,630 ... 24,700 



6,680 ... 27,610 



7,800 ... 28,870 



8,810 ... 29,350 



9, 500 ... 30, 200 



9,780 ... 30,65;'o 



10,360 ... 30,830 



10,840 ... 31,370 



11,180 ... 31,560 



Field 

 round 



per sq. cm. 

 3.900 

 6,400 

 7,710 

 8,080 

 9,210 

 9,700 

 10,610 



the metal 



19,660 



21,930 



, 22,830 



■ 23.520 



. 24,580 



. 24,900 



. 25,600 



1-98 

 4 "04 

 5-8i 

 7-60 

 ii-o 



13-5 

 i6-2 

 21-6 

 26 '8 



Current in 



field magnets, 



amperes 



.. 1-97 .. 



•• 375 ■■ 



.. 538 .. 



.. 7-08 .. 



.. 1315 .. 



.. i6-9 .. 

 .. 22-6 



£ — o utside li eld 

 4ir 



1250 

 1240 

 1200 

 1230 

 1220 

 I2IO 

 1190 



5-04 



3 '42 

 2 '96 

 2 '91 



2-67 

 2-57 

 2-46 



The magnetic force within the metal (ijj) differs from the 

 field in the .■■urrounding space by an amount which cannot be 

 estimated without a knowledge of the distribution of free 

 magnetism on the pole-pieces and conical faces of the bobbin. 

 It appears probable that, with the dimensions of the various 

 parts used in these experiments, the magnetic force within the 

 metal is less, but not very greatly less, than the outside and 

 closely neighbouring field. In the absence of any exact know- 

 ledge of J5, it is interesting to examine the relation of tJ to the 

 outside field. Thus, (ti- outside fieId)/4Tr gives a quantity 

 which is probably not much less than the intensity of magnetism, 

 jjj. The values of this quantity and also of the ratio U/outside 

 field for Lowmoor wrought-iron and cast-iron are stated in the 

 tables above. 



Curves are given showing the relation of .B to 13/outside field 

 for Lowmoor iron and for cast-iron, in the manner introduced 

 by Rowland for showing the relation of ii to /u (the permea- 

 bility). The curves have the same kind of inflection that a 

 curve of )i. and U begins to have when the magnetising force is 

 raised sufficiently high. The range through which the permea- 

 bility of iron may vary is well shown by comparing the values 



